276 
G. R. BISBY 
plasm. In this figure the protoplasm has shrunken slightly away from 
the sporangium wall. This shrinkage is undoubtedly artificial. The 
lime knot is nearly filled with cystolithic granules of calcium carbonate; 
in this instance, the acids of the fixative apparently did not penetrate 
sufficiently to dissolve out the lime. In the figures following (figures 
2-4, 11) representing succeeding stages, the calcium carbonate ap- 
pears to have been largely removed and only a small amount of sub- 
stance remains in the lime knots. In young stages, the tubular spaces 
which extend through the piotoplasm and which define the location 
of future capillitial threads, usually appear empty and are rendered 
visible only by very careful focusing. They are usually of a very 
narrow diameter; most of the mature threads, for that matter, are 
also much attenuated. The capillitial space in the center of figure i 
is, however, slightly wider than usual and contains some stainable 
substance, possibly lime. The sporangial wall in this and certain 
other figures presents an altered appearance due to the removal of the 
lime naturally present. Two aggregations are shown, however, in the 
wall of figure i which are interpreted as areas in which lime has been 
present. 
Figure 2 represents a somewhat later stage of capillitium formation. 
The protoplasm is seen to be in contact with the forming thread, a 
plasma membrane being evident only where the lime knot, from which 
the contents have been largely dissolved, has shrunken away. The 
sporangial wall in this figure is rather thick, and its inner surface is 
in direct continuity with the walls of the thread, while the thread 
itself is in turn continuous with the walls of the lime knot. This 
capillitial thread in figure 2 is clearly hollow save for an attenuated 
portion near the periphery of the sporangium, where no lumen is 
evident. 
The threads are by no means all formed simultaneously. Spaces 
without deposition may be found in sections showing capillitial threads 
which are nearly mature. 
Figure 3 shows similarly the continuity of the thickened walls of 
the lime knot and those of the threads. In this case a more advanced 
stage is figured, for the bounding walls are much thicker than in the 
previous figure. Shrinkage of the lime knot has not occurred, although 
its contents have been largely removed. 
In figure 4 the hollowness of the narrow thread is not apparent. 
Here is shown further the connection of the capillitial threads and the 
